Skip to content
Open access

Sustainable MnO₂/Biochar Nanocomposite from Pomegranate Peel for Rhodamine B Removal: Statistical Optimization and Machine Learning

Sep 2026 · Journal of Environmental Nanotechnology · 0 citations

Abstract

This study developed an MnO₂/biochar nanocomposite derived from pomegranate peel and investigated its performance for removing Rhodamine B from aqueous media. The prepared adsorbent was characterized using FTIR, BET, UV–Vis, PXRD, and FESEM analyses to determine its functional groups, porous characteristics, optical behaviour, crystalline phases, and surface morphology. A Taguchi L27 design, signal-to-noise analysis, and ANOVA were employed to examine the effects of contact time, adsorbent dosage, solution pH, and initial dye concentration. MEREC weighting and MABAC ranking were subsequently integrated for simultaneous optimization of removal efficiency and adsorption capacity. Taguchi analysis predicted the preferred factor-level combination as 60 min, pH 10, 0.05 g dosage, and 75 mg L⁻¹ concentration, whereas MABAC identified Run 24 (60 min, pH 7, 0.05 g, and 75 mg L⁻¹) as the best condition among the experimentally tested L27 alternatives, yielding 74.50% removal and 85.53 mg g⁻¹ adsorption capacity. The engineered-input ANN–GA models produced cross-validated R² values of 0.8488 and 0.9173 for removal efficiency and adsorption capacity, respectively. However, quadratic regression achieved higher leave-one-out R² values of 0.9265 and 0.9777, indicating better generalization than ANN–GA for the limited L27 dataset. The proposed framework demonstrates the combined application of statistical design, objective weighting, multi-criteria ranking, and machine learning for evaluating Rhodamine B adsorption by the MnO₂/biochar nanocomposite.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.